1. Field of the Invention
The present invention relates to a computer program product, system, and method for archiving data sets in a volume in a primary storage in a volume image copy of the volume in a secondary storage.
2. Description of the Related Art
An information life management (ILM) system manages a data set life cycle through creation until deletion. An ILM system may determine certain data sets or files to migrate to archival storage based on an archive policy, such as the data set not being accessed for a predetermined period of time. If the data set later needs to be accessed, then it may be retrieved from archival storage to the primary storage from where it can be accessed again.
Automated, policy-based data ILM operations are performed at the file level. Storage software performs this function by applying policy criteria to each file based on the file's various metadata stored in catalogs, file structures, etc. The ILM operations including the archiving of the file utilizes host system processor resources to apply the policy criteria, compress files to migrate, and perform the migration operation.
In the current art, users of the International Business Machines (“IBM®”) DB2® database software have archived database objects when copying the objects to tape. With this technique, after all of the object level copies have been created on disk, the user invokes a hierarchical storage management operation, with the DFSMShsm function to archive all of the object level copies onto tape via a MIGRATE function. This function processes all of the objects at the file level, compresses the data onto tape and then updates the catalog entries for the data object to indicate that the objects have been archived. When the object needs to be retrieved, the file system intercepts any reference to the data and automatically returns the image copy back to disk so that it can be processed. (IBM and DB2 are registered trademarks of IBM in the United States and other countries).